In the current optical communications field, an optical phase modulator represented by a QPSK modulator has become a mainstream product in the industry. After the QPSK modulator modulates an electrical signal to an optical signal, an output modulated optical signal (generally a QPSK signal) usually has two mutually conjugate possibilities, for example, I+jQ (or −I−jQ) and I−jQ (or −I+jQ), where I is an in-phase signal and Q is a quadrature signal.
A conventional QPSK signal conjugate relationship identification method, that is, a conventional method for identifying which of the two possibilities a QPSK signal belongs to, includes: performing a coherent detection of a QPSK signal by using an ICR (Integrated Coherent Receiver) and an LO (Local Oscillator), to obtain an optical signal that has undergone coherent optical interference, so as to convert phase information of the QPSK signal into optical intensity information; using a PD (Photo Diode) and a high-speed ADC (Analog-to-Digital Converter) to perform optical-to-electrical conversion and analog-to-digital conversion on the interfered optical signal, so that the interfered optical signal is converted into a digital signal; comparing the obtained digital signal with a pulse signal to be modulated, and using a PDA (Phase Detection Algorithm) to perform phase determining on the QPSK signal to obtain the phase of the QPSK signal, and to further learn a conjugate relationship of the QPSK signal.
The conventional algorithm (PDA) to perform phase determining is rather complex, and a requirement for an analog-to-digital conversion speed of the signal that has undergone the optical-to-electrical conversion is high in the prior art, which causes high implementation costs. In addition, the components (such as the ICR and the LO) used in the foregoing method are large in size, and therefore are unsuitable to be applied in an optical module in which layout space is limited.